Proceedings of the IEEE 2017 International Conference on Computing Methodologies and Communication

(ICCMC)

Implementation of Image Processing for Detection

of Brain Tumors
Miss. Shrutika Santosh Prof. Akshata Raut Prof. Swati Kulkarni
Hunnur Assistant Professor Assistant Professor
Department of EXTC, Department of EXTC, Department of EXTC,
MGM’s College of MGM’s College of MGM’s College of
Engineering and Engineering and Technology, Engineering and Technology,
Technology, Navi Mumbai Navi Mumbai, India Navi Mumbai, India
shrutikahunnur@gmail.com akshata.raut1578@gmail.com swatikulkarni494@gmail.com

Abstract -Processing of magnetic resonance images (MRI) is
one among the parts of the image processing in medical field,
which is the most emerging field from past few days. The tumor
detection is often a preliminary phase. This paper describes the
detection of the brain tumor by thresholding method. The
proposed technique can be efficiently applied to detect and
extract the brain tumor from MRI images obtained from patient
data base. It proves to be handy tool for the physicians working
in this field.
Keywords - Grayscale image, Brain Tumor, Magnetic
Resonance Imaging (MRI), Edge Detection (Using sobel operator),
Filtering, Thresholding, Shrinking operation on image
I. INTRODUCTION
Digital Image Processing is the field which refers to the
processing of digital images. A digital image is composed of
finite number of elements called as picture elements/image
elements/pixels, each of which has a particular location and
intensity value.
In the medical field for detection and visualization of
details in the internal structure of the body MRI is commonly
used. It is basically used to detect the differences in the body
tissues which is considered to be the better technique as
compared to computed tomography [1], [5], [7]. Thus, this
technique become a special technique especially for the brain
tumor detection and cancer imaging [11], [17], [18]. Basically
for comparison purpose, CT uses ionizing radiation while MRI
uses strong magnetic field to align the nuclear magnetization
that follows by changes the alignment of the magnetization by
radio frequencies that can be detected by the scanner. The
signal produced can be further processed later to gain extra
information of the body [1].
Brain tumor is a mass of tissue in which cells grow and
multiply in an uncontrollable manner. These brain tumors may
be embedded in the regions of the brain that makes the
sensitive functioning of the body to be disabled. Tumor
location and its vigorous spreading capacity makes treatment
very complex and risky.
Tumors can be classified into three types:
1) Benign
2) Premalignant
3) Malignant
Benign tumors are those which are incapable of abrupt
expanding and affecting the other healthy brain tissues.
Premalignant tumor is a pre-cancerous stage, if not treated
properly it may lead to cancers. It is often considered as a
disease. Malignant tumor grows rapidly with time an
ultimately leads to death of patient. Malignant is a medical
term describing a sever growth of a disease.
II. METHODOLOGY
The proposed work is based mainly on thresholding,
application of morphological operations and extraction of the
tumor region for further analysis. Thresholding technique is
the application of ‘T’ which is a constant over an image.
When we apply the threshold constant ‘T’ over an entire
image, the process is called as global thresholding whereas
when we change the value of ‘T’ over an image, that process
is called as variable thresholding. The image of the brain is
obtained from the MRI scanning. The experiment has been
implemented using MATLAB R2014a.

The body is made up of many cells which have their own

special function. Most of the cells in the body grow and divide
to form a new cell of the same kind as they are needed for the
proper functioning of the human body. When these cells lose
control and grow in an uncontrollable way, it gives rise to a
mass of unwanted tissue forming a tumor.

978-1-5090-4890-8/17/$31.00 ©2017 IEEE 717

 Proceedings of the IEEE 2017 International Conference on Computing Methodologies and Communication
(ICCMC)

FLOW CHART: statistics filters are whose response is based on ordering

(ranking) the pixels contained in the image area encompassed
by the filter, and then replacing the value of the center pixel
with the value determined by the ranking result. Median filter is
the best-known example, which as its name implies, replaces
the value of a pixel by the median of the gray levels in the
neighborhood of that pixel (the original value of the pixel is
included in the computation of the median) [13]. Median filters
are popular because for certain types of random noise they
provide excellent noise-reduction capabilities with considerably
less blurring than linear smoothing filters of similar size [3].

D. Thresholding of the image

Thresholding is used to extract the objects from the
background by selecting a threshold value T. Any point (x, y) in
the image at which f(x, y)>T is called an object point, otherwise
the point is called a background point. When T constant is
applicable over entire image, the process is known as global
thresholding. When the value of T changes over an image, we
use the term variable thresholding or sometimes referred to as
local regional thresholding. T is a variable, if it depends on the
spatial coordinates (x, y) themselves then variable thresholding
is often referred to as dynamic or adaptive thresholding [3].

E. Apply morphological operations

Morphological operation involves dilation and erosion.
Dilation combines two sets using vector addition. Erosion
combines two sets using vector subtraction and is the dual
operator of dilation [2], [9], [12]. Both are not invertible
transforms. Structural elements are small sets or sub-images
used to probe an image under study for properties of interest.
A. Obtaining the image:
The MRI image can be obtained from the patient data base on F. Tumor region is detected
the computer when the person undergoes the MRI scanning. Tumor region detection is done by using thresholding
Usually MRI images looks like a black and white images [14], technique and then by applying morphological operation on the
[15], [16]. The illusion of gray shading in a black and white image.
image is obtained by rendering the image as a grid of black dots
on a white background (or vice versa), with the sizes of the G. Shrinking of the image
individual dots determining the apparent lightness of the gray in
Shrinking refers to decrease the quantity of pixels. When we
their vicinity [4], [8], [6].
apply shrinking operation on an image, we are basically
removing the unwanted details that are present in the image.
B. Converting into Grayscale image
Grayscale images are the images that contain only H. Calculate area of tumor region
brightness information. Each pixel value in a grayscale image Calculation of area is one of the efficient parameter in
corresponds to an amount or quantity or intensity of light. In a detecting the stage of tumor patient.
grayscale image, brightness graduation can be differentiated
and each pixel is represented by a byte or word, the value of
which represents the light intensity at that point in the image I. Display stage of tumor patient
[20]. An Stage 0: carcinoma in situ (DCIS or LCIS) – the tumor has not
8-bit image will have a brightness variation from 0 to 255 spread past the ducts or lobules
where Stage I: The tumor is small (less than 2 cm) and well localized.
‘0’ represents black and ‘255’ represents white. Stage II: The size of tumor ranges between 2 cm and 4 cm.
Stage III: The size of tumor has exceeded 4 cm.
C. Filtering of the image Stage IV: The cancer has spread to other organs in your body,
Image which is applied to the high pass filter to remove the such as your bones, lungs, liver or brain.
noise present so as to improve the image quality and to reduce
the miscalculation. with the help of the median filter, the image
is again filtered so as to remove the impulse noise. As the value
of each pixel obtained by the median of the neighboring pixel, III. SIMULATION RESULTS
median filter is less sensitive filter is less sensitive than. Order-

978-1-5090-4890-8/17/$31.00 ©2017 IEEE 718

 Proceedings of the IEEE 2017 International Conference on Computing Methodologies and Communication
(ICCMC)

An image must be captured by source or camera and spatial domain. For removing the noise content in the image
converted into manageable entity before any video or image and for sharpening of image high pass filter is used. The output
processing is done on that image. This is the process known obtained is as shown in Figure 3.
as image acquisition. The image acquisition process consists
of three steps; energy reflected from the object of interest, an
optical system which focuses the energy and finally a sensor
which measures the amount of energy. Hence for the
processing we go for digital images. Take the MRI image of
brain as input as shown in Figure 1.

Figure3: High pass filtered image

Median filter is used to enhance the quality of the image. Noise

reduction is a typical pre-processing step to improve the results
of later processing (for example, edge detection on an image).
Median filtered output is as shown in Figure 4.

Figure1: The input image in .jpg format [10]

The grayscale 8 bit image has the intensity values ranging

from 0 to 255.The intensity levels depend upon the bits of
image. It can be calculated by using 2^n, where n represents
the number of bits. Gray level images are used as opposed to
color, not because of the "format" of gray level images but
because the inherent complexity of gray level images is lower
than that of color images. E.g. one can talk about brightness,
contrast, edges, shape, contours, texture, perspective,
shadows, and so on without addressing color. After presenting
a gray level model/ method, in most cases it can be afterwards
be extended to color images. Figure4: Median filtered image

After removing the noise and enhancing the quality of the

image, edge is detected using sobel operator as shown in Figure
5. The Sobel operator performs a 2-D spatial gradient
measurement on an image and so emphasizes regions of high
spatial frequency that correspond to edges. The approximate
absolute gradient magnitude at each point in an input grayscale
image can be found by sobel operator. The major difference
with Prewitt operator is that in sobel operator the coefficients of
masks are not fixed and they can be adjusted according to our
requirement unless they do not violate any property of
derivative masks.

Figure 2: Conversion of input image into grayscale image

In image processing filters are mainly used to suppress either

the high frequencies in the image, i.e. smoothing the image, or
the low frequencies. An image can be filtered either in the
frequency or in the spatial domain. In contrast to the frequency
domain, it is possible to implement non-linear filters in the

978-1-5090-4890-8/17/$31.00 ©2017 IEEE 719

 Proceedings of the IEEE 2017 International Conference on Computing Methodologies and Communication
(ICCMC)

A histogram is a graphical presentation of substantial

information that are grouped into data sets, for which
frequencies or occurrences are counted [19]. Histograms are
used to summarize large data sets graphically, compare process
results with specification limits, communicate information
graphically, used as a tool to assist in decision making. A
histogram can be used to organize and display the data in a
more userfriendly format. A Histogram will make it easy to see
where the majority of values falls in a measurement scale, and
how much variation there is. Figure8 represents histogram of
the grayscale image.

Figure5: Edge detection using sobel operator

Binary images may contain numerous imperfections in

particular, the binary regions produced by simple thresholding
are distorted by noise and texture. Morphological image
processing pursues the goals of removing these imperfections
by accounting for the form and structure of the image. These
techniques can be extended to grayscale images. Morphological
techniques probe an image with a small shape or template or
unit called as structuring element. The structuring element is
positioned at all possible locations in the image and it is
compared with the corresponding neighbourhood of pixels.
Dilation and Erosion are the two morphological operations used
as per algorithm. The outputs obtained are shown in Figure 6
and Figure 7.
Figure8: Histogram of the grayscale image

The purpose of thresholding is to extract those pixels from

some image which represent an object (either text or other line
image data such as graphs, maps). Though the information is
binary the pixels represent a range of intensities. Thus the
objective of binarization is to mark pixels that belong to true
foreground regions with a single intensity and background
regions with different intensities. By using thresholding
method, final tumor region is obtained and is as shown in
Figure 9.

Figure6: Morphological operation- Dilation of image

Figure9: Tumor region is detected

Not all of our images are the exact size we need them to be, so
it's important to understand how to properly resize an image
and how resizing works. When an image is resized, its pixel
information is changed. For example, an image is reduced in
Figure7: Morphological operation-Erosion of image
size, any unneeded pixel information will be discarded by the

978-1-5090-4890-8/17/$31.00 ©2017 IEEE 720

 Proceedings of the IEEE 2017 International Conference on Computing Methodologies and Communication
(ICCMC)

photo editor (Photoshop). Both image resizing and image

resampling are different concepts, the difference between both
is

that image resizing helps in changing the size at which the Table 1: Calculation of Elapsed Time
image will print without changing the number of pixels in the
image, instead the pixels are printed further apart or closer Sr. Steps Elapsed time
together. Resizing (scaling) an image will not affect screen No. (In seconds)
display. Hence we use image resizing. The shrinking of image
is shown in Figure 10. 1. Converting input image into 1.614790
grayscale image

2. Applying high pass filter 1.701474

3. Application of median filter 2.044118

Figure10: Shrinking of image

4. Edge detection using sobel 2.375892

operator
5. Thresholding operation 5.221456

6. Morphological operations 1.819508

7. Shrinking operation on image 6.345492

Figure11: Area of tumor region and display stage of tumor patient

8. Calculate area of tumor region 5.503753

The area is the number of on pixels in the image i.e. in Figure10 9. Displaying stage of tumor 4.966152
the on pixels are those having intensity level 255 (white) which patient
shows the tumor region. Area of the tumor region is calculated,
with the help of it the stage of tumor patient is displayed as
The elapsed time for entire program is 9.149068 seconds.
shown in Figure 11.
Elapsed time for each step is mentioned in Table 1.

IV. CONCLUSION
This paper describes brain tumor detection by using
thresholding algorithm and describes the comparative study
about the tumor detection. Achieved results are shown in
upper section which shows the efficient tumor detection and
also finding the boundary extraction of tumor by using sobel
edge detection operator. Size and stage of tumor is described.
MRI images are best suitable for brain tumor detection. In this
study Digital Image Processing Techniques are important for
brain tumor detection by MRI images.

V. FUTURE WORK
The future work is to extend the proposed algorithm for color
images as well as identification of the type of the tumor.

978-1-5090-4890-8/17/$31.00 ©2017 IEEE 721

 Proceedings of the IEEE 2017 International Conference on Computing Methodologies and Communication
(ICCMC)

REFERENCES
[1] A. Mustaqeem. A. Javed, T. Fatima, “An Efficient Brain Tumor
Detection Algorithm Using Watershed & Thresholding Based
Segmentation”, International Journal of Image, Graphics and Signal
Processing, Vol.4, No.10, 2012, pp34-39.
[2] N. Liu, M. Lu, “Computational Geometry, Graphs and Applications”,
Lecture Notes in Computer Science, 2010.
[3] Gonzales and Woods, “Digital Image Processing”, Pearson Education,
India, Third Edition. pp1-774.
[4] Subhranil Koley and Aurpan Majumdar, “Brain MRI Segmentation for
Tumor Detection using Cohesion based Self Merging Algorithm” 486-
2/111$26.00 ©2011 IEEE.
[5] M. Usman Akram, Anam Usman, “Computer Aided System for Brain
Tumor Detection and Segmentation” 978-1-61284-941-6/11/$26.00
©2011 IEEE.
[6] Nandha Gopal, Dr. M. Karnan, “Diagnose Brain Tumor Through MRI
Using Image Processing Clustering Algorithms such as Fuzzy C Means
Along With Intelligent Optimization Techniques”, 978-142445967-
4/10/$26.00 ©2010 IEEE .
[7] Shen, W.A. Sandham and M.H. Granat, “Preprocessing and
Segmentation of brain magnetic resonance images”, Proc. of the 4th
Annual Biomedicine, UK, pp. 149152,2003. IEEE Conf. on
Information Technology Applications
[8] J. Vijay, J. Subhashini, “An Efficient Brain Tumor Detection
Methodology Using K-Means Clustering Algorithm”, in Int. Conf. on
Communication and Signal Processing, 2013, IEEE.
[9] Sindhushri K.S, Mrs, Manjula T.R, K.Ramesha, “Detection and 3D
reconstruction of Brain Tumor from Brain MRI images”, IJRET, vol2,
no8, pp528-534, 2013.
[10] Image
https://www.google.co.in/url?sa=i&rct=j&q=&esrc=s&source=images
& cd=&ved=0ahUKE
[11] Saif .D. Salman and Ahmed A. Bahrani “Segmentation of tumor tissue
in gray medical images using watershed transformation method”, Intl.
Journal of Advancements in Computing Technology, Vol. 2, No.
4,pp123-127,OCT 2010.
[12] Mrs. Mamata .S. Kalas, “An Artificial Neural Network for Detection of
Biological Early Brain Cancer,” Intl. Journal of Computer
Applications, Vol. 1, No. 6, pp-17-23, 2010.
[13] Dr. H. B. Kekre, Ms. Tanuja K. Sarode and Ms. Saylee M. Gharge,
“Detection and Demarcation of Tumor using Vector Quantization in
MRI images”, Dr. H.B. Kekre. International Journal of
Engineering Science and Technology Vol.(2), 2009, 59-66
[14] M.C. Jobin Christ, R.M.S Parvathi, “Segmentation of Medical Image
using Clustering and Watershed Algorithms”, American Journal of
Applied Sciences, vol. 8, pp 1349-1352, 2011.
[15] Chang Wen Chen, Jiebo Luo, Kevin .J. Parker, “Image Segmentation
via Adaptive K- Mean Clustering and Knowledge- Based
Morphological Operations with Biomedical Applications”, IEEE
Trans. Image Process, vol.7, no.12, pp1673-1683, 1998.
[16] J. Selvakumar, A. Lakshmi, T. Arivoli, “Brain Tumor Segmentation
And its Area Calculation in Brain MRI images Using K- Mean
Clustering and Fuzzy C- Mean Algorithm”, Int. Conf. On Advances in
Engineering, Science and Management (ICAESM), 2012 IEEE.
[17] Ananda, Reshmi .S. Tessamma, Thomas, “Automatic Segmentation
Framework for Primary Tumors From Brain MRIs Using
Morphological Filtering techniques”, in 5th Int. Conf. on Biomedical
Engineering and Informatics, 2012,IEEE.
[18] R. B. Dubey, M. Hanmandlu, Shantaram Vasikarla, “Evaluation of
three methods for MRI brain tumor segmentation”, IEEE computer
society, ITNG.2011.92
[19] Senthilkumaran N, Thimmiaraja J, “Histogram equalization for image
enhancement using MRI brain images”, IEEE CPS, WCCCT.2014.45
[20] R. Preetha, G. Suresh, “Performance Analysis of Fuzzy C Means
Algorithm in Automated Detection of Brain Tumor”, IEEE, 2014.

978-1-5090-4890-8/17/$31.00 ©2017 IEEE 722